CT Ratio Test Results Distorted By Connected Cables 1

[Ceast]

Who can tell my why we were seeing 5.0% ratio errors on new ABB SCV CT's with load cables connected? When we disconnected the cables the error was 0.1%. BTW the test set is a Vanguard EZCT.

 

[electricpete]

Does "load" cable refer to secondary circuit?

If so first thought that comes to mind is how much impedance is presented by those cables. Trying to push too much current through too much secondary impedance of course pushes you into saturation where you see a ratio error.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[Ceast]

Primary cables. This is on a 2400 Vac system. I believe the power cables are 500MCM. We also disconnected the CT secondaries and went directly on the terminals. Same result. It wasn't until we dropped the load cables off the bus that they ratio-ed correctly.

Funny, we tested many CT's in this facility and the CT's in this line-up are the only units that reacted this way? The other CT's are older GE units that have been in-service for 30 years. They all ratio-ed > 1.0%.

 

[Ceast]

BTW the older GE CT's were tested with all power cables connected.

Regards

 

[waross]

Did the cables in any way complete a return path so that the test set could induce a flow of current?

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Ceast]

The line-up is 2 mains and 4 feeder breakers. All new gear. It's a new install. Of the 6 breakers in the line up 4 are terminated with power cables. When we tested the CT's on the 2 feeders with no cables landed the results were within spec. Shortly there after I pulled a cable from one of the other feeders and we retested the CT and it was perfect. From 5.0% error to 0.3% w/o the cable.

Nothing making sense.

 

[ScottyUK]

Sounds like you have a path for primary current through those cables that you've overlooked so far. Circuit earth and/or busbar earth at the local and/or remote end? Drain earths? If the installation is wrong then maybe armour / shield currents are another possibility?


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[waross]

I agree with Scotty. You must will be inducing a voltage in any cables passing through the CT under test. If there is a return path allowing the induced voltage to cause a current flow it will skew your test results.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[scottf]

Yep...very likely not a CT problem, but a test problem with performing the test with the CT's primary connected.

 

[davidbeach]

I'll toss in my vote for a current loop on the primary also. You need to find that current loop first thing, it probably indicates other problems.

 

[oldfieldguy]

There's a good problem...

A current loop on the primary, though, is usually detectable as a downward change in saturation voltage. You DID check saturation voltage, did you not?



old field guy

 

[Ceast]

Scotty is correct. My bad I should have recognized the fact that the cables were grounded at the opposite end!?!?!? Doh

I looked over the circuit again this am and there it was, grounded cables. Lesson learned.

OFG, the saturation voltage was about 130 volts. I suppose the ground loop could influence the saturation voltage. I'm not really clear on that. Explain.

Thanks

 

[oldfieldguy]

Ceast:

If there is a current path, a loop through the primary, as voltage is raised on the secondary to perform saturation tests, as soon as the secondary voltage is sufficient to flow current through the inadvertent primary path, current will flow. On the secondary winding, this will appear as if the core has reached saturation. This value will be noticeably lower than the actual saturation voltage.

That is why saturation tests of installed CT's are important. I have found these unintended paths on installed CT's that have resulted in mis-operation on out-of-zone faults in differential and distance schemes.

old field guy

 

[stevenal]

My experience with a primary current loop was discussed in thread238-204742.

 

[Ceast]

Stevenal, that is an extensive read! I need to read it again. Did I read, that stacked CT's can influence each other during testing? Should the none tested CT be shorted while the other is under test?

 

[stevenal]

Ceast,

With the primary current loop in place, the stacked CTs did influence each other. Normally, they should work independently.

As a general rule, I would keep the untested CT shorted for safety.

 

[scottf]

I concur with Stevenal.

Regardless of the question of mutual influence, CTs not under direct test should always be shorted. That's a basic safety precaution.

 

[busbar]

Sorry I’m a little late in commenting, but it sounds like the situation at hand may be describead in IEEE C57.13.1, IEEE Guide for Field Testing of Relaying Current Transformers, ¶ 11.3 Intercore Coupling Check.